DESCRIPTION: We will continue our effort to understand the genesis of magnetoencephalographic (MEG) signals in terms of the modern concepts of dendritic and somatic electrophysiology in order to help interpret MEG signals from the human brain in disease and health. Our work has characterized the MEG signals produced by guinea pig hippocampal slicers. Intracellular and extracellular field potential data was well as MEG signals were obtained after systematically blocking the various ligand-gated and voltage-and calcium-sensitive channels. This has brought our research to a new stage where it is possible to make some quantitative comparisons between the three types of signals and those generated by a mathematical model (Traub model) of CA3. The model we have used contains 100 excitatory cells and 20 inhibitory cells, each excitatory cell having two types of excitatory and inhibitory receptors and six different voltage- and calcium-sensitive conductances. We will Extend the Traub model to predict no only intracellular potentials, but also field potentials and MEG signals. Our modeling work has shown that the Traub model can be extended to make some quantitatively accurate predictions. We will first apply the extended model to account for the three sets of data collected thus far and identify the aspects of the data that are well explained by the model and those that are not. The comparisons will be used to revise the model to better account for the three data sets simultaneously. The revision will, for example include changes in the distribution of the channel densities and receptor site along the dendrites and addition of new channels. As the model is improved, it should be possible to infer with increasing levels of confidence the role of different types of currents in generating the MEG signals. Experimental analyses of this issue are often ambiguous even with the use of selective channel blockers. A mathematical model helps us clarity their roles since the currents can be separately calculated. On the basis of our work, we expect that the calcium conductance plays an important role in generating MEG signals and field potentials. The comparison will also be used as the basis for specifying a new set of experiments that will best characterize the role of individual channels in generating MEG signals. In one series of study, we will characterize the signals generated with synaptic transmissions blocked and the pyramidal cells directly excited. We will also combine the model and experiments to understand the MEG signals underlying spontaneous activities such as the gamma-oscillations discovered in the hippocampal slices. Our preliminary study indicates that such oscillation scan be recorded in our preparations.